I don’t have any errors in my `apollo` pod logs, b...
# ask-community
w
I don’t have any errors in my
apollo
pod logs, but I see this in the Agent logs:
Copy code
urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host='prefect-server-initial-apollo.prefect', port=4200): Max retries exceeded with url: /graphql (Caused by NewConnectionError(': Failed to establish a new connection: [Errno 111] Connection refused'))
(prefect is my namespace, prefect-server-initial-apollo is the name of the service)
n
Hi @Wilson Bilkovich - if you ping the apollo pod directly do you get a response?
w
Let me check; thanks for the reply.
Copy code
% telnet prefect-server-initial-apollo 4200
Trying 127.1.27.7...
Connected to prefect-server-initial-apollo.
Escape character is '^]'.
GET /graphql

HTTP/1.1 400 Bad Request
X-Powered-By: Express
Access-Control-Allow-Origin: *
Content-Type: text/html; charset=utf-8
Content-Length: 18
ETag: W/"12-7JEJwpG8g89ii7CR/6hhfN27Q+k"
Date: Fri, 27 Aug 2021 15:54:51 GMT
Connection: close

GET query missing.Connection closed by foreign host.
n
And this agent is the kubernetes agent started with the
--set agent.enabled=true
flag?
w
It is, yeah; using the default tenant also created by the Helm chart
n
Are you using any networking policies in your cluster?
That might be blocking requests?
w
I haven’t added anything, let me double-check if there are any. This is Amazon EKS and it does a few things by default.
Nope, no network policies. Also, I can connect from the agent to apollo when I exec into it:
Copy code
root@prefect-server-initial-agent-5f76b4fd69-vn74p:/# telnet prefect-server-initial-apollo 4200
Trying 172.20.65.234...
Connected to prefect-server-initial-apollo.prefect.svc.cluster.local.
Escape character is '^]'.
GET /graphql?query=foo

HTTP/1.1 400 Bad Request
X-Powered-By: Express
Access-Control-Allow-Origin: *
Content-Type: application/json; charset=utf-8
Content-Length: 146
ETag: W/"92-sfDAzDmlWzD691zaMG6MP1fVCE8"
Date: Fri, 27 Aug 2021 17:17:41 GMT
Connection: close

{"errors":[{"message":"Syntax Error: Unexpected Name \"foo\"","locations":[{"line":1,"column":1}],"extensions":{"code":"GRAPHQL_PARSE_FAILED"}}]}
Connection closed by foreign host.
n
Hm, I'm not entirely sure how to proceed then, if you can query apollo from the rest of the cluster (i'm assuming the UI works) and you can query apollo from the agent container as well I'm at a bit of a loss. Can you start the agent manually ?
w
Hmm, maybe I am just chasing a ghost now, because there’s nothing unhappy in the pod logs for the current instance of the Agent.
n
Ah so it's working now? Maybe a network blip?
w
I’m trying to prove that to myself via Kibana, which I am growing to dislike..
🤔 1
Apparently these logs are from some earlier incarnation of my cluster, because I can’t reproduce it. Thanks for the help, I’ll just assume this was something silly I did in Kubernetes.
1
n
Glad you got it resolved!